This page summarizes a report in CDC's MMWR by staff of the Minnesota Department of Health's newborn hearing program and colleagues, covering infants born in Minnesota during 2012–2016.
The benchmark
Guidelines from the Joint Committee on Infant Hearing for Early Hearing Detection and Intervention (EHDI) say that any infant who does not pass newborn hearing screening should have diagnostic audiologic testing that identifies hearing loss by age 3 months.
EHDI programs have brought the average age at identification down substantially, and national data show excellent screening rates. Even so, in 2016 only 75.9% of infants who did not pass screening had documented definitive diagnostic testing by age 3 months. Of those found to have permanent hearing loss, only 67.3% were enrolled in Early Intervention services by the benchmark of age 6 months.
That gap matters. Entering Early Intervention earlier can improve language outcomes for infants who are deaf or hard of hearing (DHH), and a late identification can delay the referral and raise the risk of language delay. Few studies had asked what predicts late identification among infants already found to be DHH.
The study
Minnesota EHDI staff analyzed deidentified records of 729 infants born in Minnesota during 2012–2016 who did not pass newborn hearing screening and were then identified as DHH. The records came from the Minnesota Department of Health's EHDI information system, which collects demographic, screening, diagnostic and intervention data from birth care providers and facilities, audiologists, public health nurses, and birth certificates through the Minnesota Office of Vital Records.
Identification counted as delayed when it came after age 3 months. The factors examined were:
- where the family lived and where the infant was born;
- the mother's race or ethnicity, country of origin, age, and education at the time of the birth;
- the main language used at home;
- birthweight;
- the infant's health insurance.
Minnesota is home to some of the largest Somali and Hmong populations in the United States, so both were analyzed as separate groups. Where a birth certificate listed more than one race for the mother, the team used the bridged category the Office of Vital Records derives with National Center for Health Statistics methods.
Prevalence ratios were estimated with a modified Poisson regression model, first for each factor alone and then adjusted for the others. Birth hospital location tracked residence so closely that it was left out of the adjusted model, which used the 686 records complete for every variable. As a program evaluation, the study was exempt from Institutional Review Board review.
Findings
222 of the 729 infants (30.4%) were identified late. Taken one at a time, residence, birthweight, home language, and the mother's race or ethnicity, country of origin, age and education, along with insurance, were all linked to delay. After adjustment:
| Factor | What the report found | Adjusted prevalence ratio (95% CI) |
|---|---|---|
| Very low birthweight (under 1,500 g) | 82% identified late; more than twice as likely to be late as infants of normal birthweight | 2.6 (2.0–3.3) |
| Moderately low birthweight (1,500–2,499 g) | significantly more likely to be late | 1.4 (1.0–1.8) |
| Mother with a high school diploma or less | 40% late, against 16% for mothers with a college degree | 1.7 (1.2–2.5) |
| Mother with some college | 36% late | 1.6 (1.2–2.3) |
| Mother under 25 | about half late; compared with mothers aged 25–34 | 1.4 (1.1–1.8) |
| Mother Hmong | 60% more likely to be late than when the mother was white | 1.6 (1.1–2.5) |
| Home outside the Twin Cities metropolitan area | more likely to be late | 1.4 (1.0–1.9) |
| Public health insurance | more likely to be late | 1.6 (1.1–2.2) |
Birthweight was the strongest factor. Nearly half of infants whose mothers were black were also identified late; their ratio, 1.5 (1.0–2.3), was among the largest but did not reach significance at p<0.05.
Possible reasons
- Socioeconomic factors are well-documented determinants of health, and several such indicators were linked to delay here. Why audiologic follow-up is harder for people with lower socioeconomic status needs more study. Two practices already look promising: working with birth and primary care providers on messaging about follow-up after a failed screen, and scheduling the next test at the moment an infant does not pass.
- Very-low-birthweight infants are at higher risk for many complications around birth, many of which appear on the Joint Committee's list of risk factors for permanent congenital, delayed-onset or progressive childhood hearing loss. They may also be too medically fragile in the Neonatal Intensive Care Unit for a timely diagnostic test.
- The Hmong finding had not been reported before, and the literature holds few hearing studies involving Hmong people. Its causes are unclear and need further study.
- Living outside the metropolitan area was linked to delay, as in earlier research. Pediatric audiology may be scarce in some nonmetropolitan regions. Tele-audiology has been piloted in Minnesota with some success, but the programs need expanding and refining.
Limitations
The authors list five:
- Vital records data are self-reported and open to reporting bias.
- Residence, language and insurance data come from audiologists and public health nurses, and may contain errors.
- Factors outside the data set, such as other health conditions, may have affected the results.
- Some comparison groups were small, making associations hard to detect.
- Delay was treated as yes or no, so the report cannot say how much longer the delay was for any group.
What could help
Late identification can delay Early Intervention, which has been shown to lead to poorer language outcomes. The report suggests:
- stronger partnerships with local public health teams that make home visits to low-birthweight infants after they leave the hospital;
- partnerships with Special Supplemental Nutrition Programs for Women, Infants, and Children, which reach low-income families, many of them publicly insured — teams that can encourage families to book diagnostic hearing tests, or book them for them;
- information for families in formats that do not depend on literacy or education, such as podcasts or online videos.
Sources
- Meyer AC, Marsolek M, Brown N, Coverstone K. "Delayed Identification of Infants Who Are Deaf or Hard of Hearing — Minnesota, 2012–2016." MMWR 69(11). https://www.cdc.gov/mmwr/volumes/69/wr/mm6911a6.htm
- The report's methods paragraph defines the outcome as identification "by 3 months of age," while its abstract and analysis define delay as identification after age 3 months; this page follows the abstract and analysis.
Licence: CC0 1.0 (public domain) · Adapted from www.cdc.gov
1
0
0
0
Comments
